viruses
The main thesis of Behe's new book, Darwin Devolves, surrounds what Behe calls "poison-pill" mutations, which gives an organism a quick fix, but which can run the risk of being incapable of utilizing future needed adaptations. In other words, breaking and blunting genes to adapt to new environments become changes that get locked in due to natural selection's tendency to root out anything but what is the 'fittest' in any environment - and this can include even beneficial mutations being rooted out due to beneficial mutations being so rare and showing up way too late to modify the adapted organism.

So, today at Phys.Org there is a PR (press release) about a study involving viruses. It turns out that even at the level of viruses, the First Rule of Adaptative Evolution applies: a broken gene ends up being beneficial to the virus, allowing it to replicate itself when it has been rendered almost unable to do so by the host's immune system.

From the PR:
But the researchers continued to culture the B1-free strain for multiple generations in the lab, then sequenced its entire genetic code to gauge how it evolved. They found that, over just a few days, the B1-free strain responded by deleting a single base pair - a fundamental component of DNA - while leaving nearly 200,000 others untouched. The seemingly miniscule loss corresponded with a 10-fold increase in the strain's otherwise stunted replication.
As usual, the experimenters are "surprised":
"We were expecting that the virus may adapt another gene to compensate," said Wiebe, associate professor of veterinary medicine and biomedical sciences. "What we found instead is that the virus adapted by inactivating another gene. It was as if, upon cutting one wire, the best way to fix the problem was to cut another wire.
Just think, if they had read Behe's new book, they would not have been surprised at all.

Update:

In his book, Behe uses evidence showing that "devolution" occurs in bacteria and in eukaryotes. Now we can add viruses to the lot. I think this only adds to and strengthens his argument for his First Rule.